Computational Heuristics for Prediction of Heat Transfer Characteristics of Quenchants

The Heat Transfer Coefficients (HTC) of Heat Flux (HF) as functions of surface temperature is the most common physical metrics used to characterizes the heat extraction ability of quenchants. Several numerical techniques have been developed to determine of HTC or HF in the last three decades. The earliest methods are based on analytical approaches and empirical formulae. Beck and Olifanov introduced optimization schemes to solve the Inverse Heat Conduction Problem. The latest research trends are focusing of the application of bio-inspired algorithms, AI methodologies including Physics Informed Neural Networks. In this paper, the reliability, accuracy, robustness and computational effectiveness of the heuristics are analyzed.